Trout hepatomaFibrosis and lymphocytosis as suppressive mechanisms in the rainbow trout Salmo gairdneri.

Trout Hepatoma : Fibrosis and Lymphocytosis as
Suppressive Mechanisms in the Rainbow Trout
(Sarimo gairdneri) '
J. H. WALES AND R. 0.SINNHUBER
Department of Food Science and Technology, Oregon State University,
Cornallis, Oregon 97331
ABSTRACT
Fibrosis and lymphocytosis are common in aflatoxin-induced
hepatoma in rainbow trout. It is suggested that these reactions may control the
growth of tumors and in some cases destroy them. Nearly all of the hepatomas
we have observed are fibrotic to some extent. The extent of the fibrosis seems to
be age dependent and in advanced cases fibrosis appears to have disorganized the
tumors to a degree which we believe is suppressive. The lymphocytosis shows a
negative correlation with the size and age of the hepatoma, being found almost
exclusively in the very "early" (small) hepatomas and preneoplastic nodules. It
is suggested that the invasion of the tumors by lymphocytes is a host defense
mechanism.
of carcinogens and cocarcinogens have
been tested under a wide variety of experimental conditions and in conjunction with
many dietary modifications (Sinnhuber et
al., '68). Large numbers of hepatomas can
be produced by feeding any nutritionally
balanced trout diet to which has been added
aflatoxin B1 at levels within the range of
4 to 20 parts per billion (ppb). The carcinogen is added to the diet as soon as possible after feeding starts. Hepatomas may
reach macroscopic size in four months although this is temperature-dependent and
in water averaging 12" C large numbers
of hepatomas are rarely obtained in less
than six months. Livers containing hepatomas are routinely fixed in Bouin's fluid,
embedded in paraffin and stained with
hematoxylin and eosin. Masson's triple
stain is occasionally used to emphasize the
connective tissue in fibrotic tumors.
In the course of this work, a very large
number of hepatomas has been examined.
Certain carcinogenic diets produce as many
MATERIALS AND METHODS
as 100 tumors in a single liver, ranging
The brood stock of Mt. Shasta "strain" from microscopic size to more than 8 cm
rainbow trout (Salmo gairdneri) main- in diameter. Unfortunately the observatained in our laboratory has been found to tions of fibrosis and lymphocytosis are diffibe more sensitive to aflatoxin-induced cult to treat statistically. Only a rough estihepatoma than any other animal tested.
Received July 17, '72. Accepted Oct. 3, '72.
1This work was supported in part by U.S.P.H.S.
Approximately 200 diet trials have been
grants ES-00263 and ES-00256.Technical paper 3220,
run in our laboratory since 1964. A number Oregon Agricultural Experiment Station.
Infiltration of hepatomas in the rainbow
trout by ductal tissue components (fibrosis) and by lymphocytes has been described
by a number of workers (Haddow and
Blake, '33; Cudkowicz and Scolari, '55;
Nigrelli and Jakowska, '61; Wood and Larson, '61; Rucker et al., '61; Hueper and
Payne, '61; Ghittino and Ceretto, '62; Ashley and Halver, '63; Scarpelli et al., '63;
Ashley, '69; Wales, '70; Ashley, '70). Although these workers have mentioned the
presence of either or both tissues in trout
hepatomas, they have not discussed their
importance in controlling the growth or
causing the regression of tumors. Hellstrom and Hellstrom ('69) have reviewed
the subject of lymphocytic antagonism to
tumor cells. The body of literature concerned with host immune reaction in neoplasia is already large and growing rapidly,
but apparently there have been no published reports of this phenomenon in trout
hepatoma.
ANAT. REC., 175: 97-106.
97
98
J. H. WALES AND R. 0. SINNHUBER
mate of incidence can be made from the
skip-sectioning technique employed.
age of such cases must be very small,
possibly 10%.
OBSERVATIONS
Lymphoc ytosis
The presence of foci of lymphocytes
within a liver normally indicates the presence of disease organisms or injured tissue
but in the livers of the trout which have
been fed carcinogenic diets the areas of
lymphocytosis are usually restricted to
hepatic nodules (figs. 9-12). Usually these
are nodules which we class as eosinophilic
or preneoplastic but occasionally they are
assumed to be small hepatomas. Foci of
lymphocytes are found more frequently in
livers of trout fed a mycotoxin plus cyclopropenoid fatty acids (CPFA) (Sinnhuber
et al., ’66). Presumably the CPFA reduces
the host defense mechanisms and a greater
number of incipient hepatomas reach a
recognizable size. These defense mechanisms, we assume, do not include lymphocytosis.
The extent to which these small nodules
are invaded by lymphocytes is variable but
usually it is but a small portion of the
entire hepatoma. Frequently we find foci
of lymphocytes in areas of the liver sections which are not clearly neoplastic or
“preneoplastic” and there may be reasonable doubt about the stimulus which has
produced this reaction. In most cases, however, the character of the liver parenchyma
being invaded by lymphocytes is that of
neoplastic tissue. In most instances the
situation is similar to that illustrated in
figures 9 and 10 where the distribution of
lymphocytes is sharply confined to the abnormal cells.
Nearly all of the neoplasms observed in
our experimental fish have been trabecular
hepatocarcinomas in the various stages of
development or evolution which they undergo. The term “hepatoma” will be used
in the broad sense to include all the stages
which the writers believe are clearly developmental forms or variations of hepatocarcinomas. Possibly the only departure
from the common parenchymal cell modification are the pseudo-bile ducts observed
in certain hepatomas. These appear to arise
within areas of fibrosis and are presumed
to be modifications of bile duct epithelium.
Fibrosis
Our tissue sections indicate that the
monocellular strands of loose connective
tissue wander from the point of origin,
usually near the center of the tumor, toward the periphery of the hepatoma. Apparently the connective tissue cells involved in this process originate from the
supporting sheath surrounding blood vessels and bile ducts. These strands become
multicellular bundles and eventually form
continuous masses of fibrous cells with
occasional islets of hepatoma cells enclosed therein (figs. 1-4). As this process
continues the center of the tumor becomes
almost entirely fibrotic with necrotic spaces
containing a semi-fluid mass. The neoplastic parenchyma becomes a shell around
the periphery of the fibrotic mass and even
this shell of hepatoma tissue may become
completely overgrown and presumably
ceases to grow (figs. 5-8). This process is
highly variable, each tumor may present a
slightly different picture. Occasionally the
fibrous strands will not terminate their outward growth at the tumor surface but will
spread along the surface developing a fibrous sheath over the tumor. If the tumor
extends to the surface of the liver it will
touch the abdominal wall or some visceral
organ such as the pyloric caeca and in such
cases the fibrous strands may continue beyond the tumor into the body wall or visceral organ forming adhesions. Very rarely
a hepatoma may contain no visible connective tissue hyperplasia but the percent-
DISCUSSION
Fibrosis
Although the term “fibrosis” is applied
to the situation which develops within most
trout hepatomas it should be made clear
that we are referring to both the loose connective tissue cells and the collagenous
fibers which accompany them. Microscopically this is quite similar to the aflatoxin-induced “bile-duct proliferation” commonly observed in duckling livers (Asplin
and Carnaghan, ’61). However, the two
types of hvperplasia seem to be dissimilar
etiologically. In the duckling, for example,
the proliferation of this connective tissue
TROUT HEPATOMA: FIBROSIS AND LYMPHOCYTOSIS
99
occurs in the non-neoplastic hepatic tissue parenchyma interlaced by strands and
whereas in the trout this hyperplasia is bundles of connective tissue. We are conrestricted to the hepatoma and only rarely vinced from abundant observations that
extends beyond the periphery of the hepa- fibrosis precedes necrosis and we hypothetoma. This difference leads to speculation size that the islets of neoplastic parenabout the stimuli involved. Obviously the chyma, isolated by fibrosis, become neneoplastic tissue is compatible with the crotic.
hyperplastic connective tissue and appears
Lymphocytosis
to stimulate its growth. On the other hand
It could be argued that disease organthe normal liver parenchyma appears to
be incompatible with the hyperplastic isms or necrosis elicit the lymphocytic reductal cells but compatible with similar sponse but these views lack support. Our
duct cells in the normal portion of the fish rarely have diseases of any kind and
liver. This resembles the homograft re- no case has yet been found in our laborajection phenomenon and it raises the ques- tory in which a liver had been invaded by
tion whether there is a significant degree any organism. In addition we f h d lymof incompatibility between the normal phocytosis in very small, presumably very
liver tissue and the hepatoma. The rela- young nodules, less than a millimeter in
tionship of normal liver parenchyma to diameter, which are simply colonies of
the various elements of the ductal system atypical cells without fibrosis or any newithin the liver indicates that there is com- crosis. Consequently we view these lympatibility but no evidence of stimulation. phoid invasions of the hepatomas as imThis condition appears to be altered when mune reactions to a foreign substance.
Our observations lead us to believe that
the parenchyma transforms from normal
there are two distinct phenomena involved.
to neoplastic.
Invariably the center of a hepatoma is The connective tissue is “permitted” or
the location of the greatest development “stimulated” to hyperplastic growth by the
of connective tissue. Apparently a small neoplastic liver parenchyma. The lymphoid
colony of hepatoma cells in its early de- tissue, on the other hand, is attracted to
velopment will eventually surround a duct. the neoplastic cells, possibly by a substance
The connective tissue element of the duct secreted by these malignant cells.
We believe that these observations lend
is stimulated to hyperplastic proIiferation
even though the hepatoma may be a milli- support to our hypothesis that fibrosis and
meter in diameter. Normally such hepa- lymphocytosis can, in appropriate circumtomas grow spherically; thus the center will stances, control the growth of, or combe the oldest portion and the area most pletely destroy hepatomas in the rainbow
likely to contain connective tissue. We trout liver.
believe that the sequence of events is that
ACKNOWLEDGEMENT
described above rather than the alternative
We
wish
to thank Mrs. L. ,T. Hunter for
proposition that necrosis begins in the
center of a tumor, due to lack of circula- her assistance in preparing the tissue for
tion, and is followed by the connective examination.
tissue invasion to occupy the space. We
LITERATURE CITED
have in our collection of trout hepatomas
Ashley,
L.
M.
1969 Experimental fish neoseveral hundred tumors ranging in diamplasia. In: Fish in Research. Neuhaus and
eter from microscopic to 80 mm. Only the
Halver, eds. Academic Press, New York-Lonlargest tumors, approximately 40 mm or
don, pp. 23-43.
1970 Pathology of fish fed ailatoxins
larger, contain areas of necrosis similar to
and other antimetabolites. In: A Symposium
those shown in figure 13. It is probably true
o n Diseases of Fishes and Shellfishes. Snieszko,
that connective tissue possibly supports the
ed. Special Pub. No. 5, Am. Fish. SOC.,Wash:
architecture of the tumor after central
ington, D. C., 366-379.
necrosis has eroded much of the interior Ashley, L. M., and J. E. Halver 1963 Multiple
metastasis of rainbow trout hepatoma. Trans.
tumor tissue but prior to this late staqe of
Am. Fish. SOC., 92: 365-371.
development the tumors are completely Asplin, F.D., and R. B. A. Carnaghan 1961 The
filled with vigorously growing neoplastic
toxicity of certain groundnut meals for poultry
100
J. H. WALES AND R. 0. SINNHUBER
with special reference to their effect on ducklings and chickens. Vet. Record, 73: 1215-1219.
Cudkowicz, G., and C. Scolari 1955 Un tomore primitivo epatico a diffusione spizootica
della trota iridea di allevamento (Salmo irideus). Tumori, 41: 524-537.
Ghittino, P., and F. Ceretto 1962 Studio s d a
eziopatogenesi dell' epatoma della trota iridea
di allevamento. Tumori, 48: 393-409.
Haddow, A., and I. Blake 1933 Neoplasms in
fish: a report of six cases with a summary of
the literature. J. Path & Bact., 36: 41-47.
Hellstrom, K. E.,and I. Hellstrom 1969 Cellular immunity against tumor antigens. Advan.
Cancer Res., 12: 167-223.
Hueper, W.C., and W. W. Payne 1961 Observations on the occurrence of hepatomas in rainbow trout. J. Nat. Cancer Inst., 27: 1123-1143.
Lee, D. J., J. N. Roehm and R. 0. Sinnhuber
1967 Effect of w3 fatty acids on the growth
rate of rainbow trout. Salmo gairdneri. J. Nutr.,
92: 93-97.
Nigrelli, R. F., and S. Jakowska 1961 Fatty
degeneration, regenerative hyperplasia and
neoplasia in the livers of rainbow trout, Salmo
gairdneri. Zoologica, 48: 49-55.
Rucker, R. R., W. T. Yasutake and H. Wolf 1961
Trout hepatoma
a preliminary report. Prog.
Fish-Cult., 23: 3-7.
Scarpelli, D. G., M. H. Greider and W. J. Frajola
1963 Observations on hepatic cell hyperplasia,
adenoma and hepatoma of rainbow trout
(Salmo gabdneri). Cancer Res., 23: 84-57.
Sinnhuber, R. O., J. H. Wales, J. L. Ayres, R. H.
Engebrecht and D. L. Amend 1968 Dietary
factors and hepatoma in rainbow trout (Salmo
g a i r d n e r i ) . I. Aflatoxins in vegetable protein
feedstuffs. J. Nat. Cancer Inst., 41: 711-718.
Sinnhuber, R. O.,J. H. Wales, R. H. Engebrecht,
W. D. Kray, et al. 1965 Aflatoxins in Cottonseed meal and hepatoma in rainbow trout. Fed.
Proc., 24: 627 (Abstract).
Sinnhuber, R. O., J. H. Wales and D. J. Lee 1966
Cyclopropenoids, cocarcinogens for aflatoxininduced hepatoma in trout. Fed. Proc., 25: 555
(Abstract ) .
Wales, J. H. 1970 Hepatoma in rainbow trout.
In: A S y m p o s i u m on-Diseases of Fishes and
Shellfishes. Sniesko, ed. Special Pub. No. 5, Am.
Fish. SOC.,Washington, D. C., pp. 351-365.
Wood, E. M., and C. P. Larson 1961 Hepatic
carcinoma in rainbow trout. Arch. Path., 71:
471-479.
-
PLATE 1
EXPLANATION O F FIGURES
1 Hepatoma of rainbow trout surrounded by normal liver. The entire
hepatoma ( 2 m m diameter) is divided by heavy strands of connective
tissue. Hematoxylin and eosin ( H & E). x 20.
2
Hepatoma (4 mm diameter) containing slender strands and masses
of connective tissue (arrow). H & E. X 20.
3
Detail of hepatoma (5 m m diameter) showing a cluster of neoylastic
parenchymal cells (arrow) surrounded by connective tissue. At this
stage of fibrosis there is no evidence of necrosis. H & E. X 128.
4 Central area of hepatoma heavily overgrown by connective tissue.
Despite the division and isolation of the hepatoma tissue there is no
evidence of necrosis. H & E. X 128.
TROUT HEPATOMA : FIBROSIS AND LYMPHOCYTOSIS
J. €
Wales
I. and R. 0.Sinnhuber
PLATE 1
101
PLATE 2
EXPLANATION OF FIGURES
5
Details of hepatoma tissue divided by fibrous strands. H & E. X 320.
6
A single isolated group of neoplastic parenchymal cells surrounded by
fibrous tissue. H & E. x 800.
7 Detail of hepatoma tissue divided by fibrous strands. H & E. X 320.
8
102
A necrotic islet of hepatoma cells entirely surrounded by connective
tissue of bile duct origin. H & E X 320.
TROUT HEPATOMA: FIBROSIS AND LYMPHOCYTOSIS
J. H. Wales and R. 0.Sinnhuber
PLATE 2
PLATE 3
EXPLANATION O F FIGURES
9
Section of trout liver containing a small hepatoma approximately
1 mm diameter. Note that within the hepatoma are several darkly
staining foci of lymphocytes. H & E. x 20.
10 Margin of darkly staining hepatoma with several areas occupied by
lymphoid tissue. H & E. x 128.
11
Central portion of hepatoma heavily invaded by lymphoid tissue.
H & E. X 128.
12 Detail of hepatoma which has been occupied by lymphoid tissue leaving a few isolated clusters of neoplastic parenchyma. H & E. x 320.
104
TROUT HEPATOMA: FIBROSIS AND LYMPHOCYTOSIS
J. H. Wales and R. 0. Sinnhuber
PLATE 3
105
TROUT HEPATOMA: FIBROSIS AND LYMPHOCYTOSIS
J. H. Wales and R. 0.Sinnhuber
PLATE 4
EXPLANATION OF FIGURE
13 Longitudinal section through the liver of a two year old rainbow trout. Note the single,
large tumor composed largely of neoplastic parenchyma. The center of this tumor contains connective tissue and vacuoles of necrotic hepatocytes. Another smaller tumor
(above) has been completely overgrown by connective tissue and its growth has presumably ceased. A fragment of normal liver (left) remains. x 2.5.
106